Sáhkku, The “Devil's Game” / Alan Borvo - Board Game Studies

Sáhkku, The “Devil’s Game” / Alan Borvo 

Until recently, the Sámit from Lappland – those we previously called “Lapps” (1) 

– used to play a board game that seems to be strangely connected with the family 

of the tâb race games, particularly widespread over the whole Arab-Muslim 

area. (2) Usually known under the name sáhkku (3), this game has the reputation as the 

“Devil’s game” because it was banned by the Laestadianists, a rigorist Lutheran church 

with a strong influence upon the Sámi people. 

As Peter Michaelsen has already pointed out, (4) it is evident that the Sámi sáhkku 

and the Danish daldøs (like its south-Norwegian counterpart called daldøsa) have a strong 

relationship – a relationship still reinforced by their relative geographical proximity. 

However, the Sámi game shows two main special features which make it different from 

the other two: first, the fact that the opponents are called ‘men’ and ‘women’; second, 

the existence of a ‘king’ which may change owners during play, and, sometimes, the further 

addition of two ‘king’s sons’. 

General principle of the game 

Basically, the game consists of two armies, each with the same number of pieces, 

moving forwards to fight each other. The board is a long shaped wooden plank with 

only three rows breadthwise, each army occupying one of the two outside rows at the 

beginning of play (fig. 1). There are generally three dice. 

Fig. 1. Sáhkku from Norwegian Finnmark as described by Friis in 1871.There are 15 

‘men’ called olbmak (1), 15 ‘women’ called galgok (2), one king called gonagas (3) and 

three 4-sided dice (4, 5 and 6). In this game the cross sign on the dice (sáhkku) gives 5 

points.The numbers on the board have been added for a better understanding of the 

circuit the pieces must follow. Usually the places are materialized by lines carved into 

the wood – a middle one lenghtwise and from 8 to 15 parallel ones breadthwise. 

The pieces move according to the values given by the dice, but none is allowed to 

start before it has been “activated” by the throw of one sáhkku, the side marked with a 

cross. Then they follow their own track: on Friis’s figure from point 1 to 16, 16 to 30,

34 

B OARD G AMES S TUDIES 4, 2001 

30 to 45, 45 to 31, 31 to 16, 16 to 30 and 30 to 1 for the 1-15 based camp. The opponents 

do the same but in symmetrical order: point 45 to 30, 30 to 16, 16 to 1, etc. (5) 

Each player uses the value given by the dice for moving one, two or three pieces, but 

the move of each piece must correspond to the exact values of one, two or three dice. It 

is possible to put several pieces on the same point, but, on the next turns, he may not 

move them all together with the value of the same die. When one piece arrives at a place 

occupied by one or several enemies, this or these are killed and removed from the board. 

The winner is the player who has killed all his enemies. 

Although the pieces must follow their track, the king may go in all four directions. 

Placed at the start in the middle of the central row (point 23), it belongs to the first 

player who reaches its place. But it may be captured by the opponent who puts one of 

his pieces on the point on which it stands. In that case, the king is not killed, but belongs 

to the player who has captured it. The player who owns the king may move it on his turn 

by using the throws given by the dice, in place of or together with one or two pieces. But 

the possession of the king does not allow any extra throw. 

When an opponent’s piece or king reaches the point just on the right of one’s nonactivated 

pieces, all of these are “frozen” until the opposite man or king goes away. But 

a non-activated piece may never be killed. 

In some local variants, as already remarked, two extra ‘king’s sons’ were used, the 

initial places of which were marked with crosses on various old boards, on both sides of 

the king’s point in the central row (points 19 and 27). Unfortunately, we know nothing 

of their role. 

Another interesting item of information which we got from Edmund Johansen (a Sea 

Sámi said to be the one who knows best the rules since he frequently played sáhkku 

when he was a child) is that the two players may decide not to play “against” but “together” 

– apparently a much more tricky affair. In that variant, used only by wise elders, 

the two armies had their pieces displayed at the beginning on the same side of the board. 

The target of the game was the same, but nobody remembers how it was played 

(Johansen 2000-2001). 

A challenge for the ethnologist 

The interest of such a game is manifold, for it raises questions whose answers may 

help to clarify certain points in our knowledge of the Sámit. Where does this game come 

from, and what outside influence does it show? Do its geographical dispersion and the 

distribution of its local variants bring anything new concerning the relationships between 

the various communities? Is what the Samit added of their own to the game related to 

their specific thoughts, social life organization and former religious beliefs? 

Solving such questions is a real challenge. The sources are scarce and the game has 

already almost died out entirely, even if people like Edmund Johansen and his wife Eli 

strive to make it come to life again, organizing occasional tournaments in their small village 

of Kunes. In Nesseby, too, a coastal city of eastern Norwegian Finnmark on the 

north bank of the Varangerfjord, a few old men have made a copy of a 1906 set on exhibition 

at the nearby Varanger Samiske Museum, but they had to reconstruct the rules

A. BORVO, SÁHKKU, THE “DEVIL' S G AME” 35 

Fig. 2. Old sáhkku set exhibited at the Varanger Samiske Museum (eastern Norwegian 

Finnmark), on long-term loan from the Ethnografisk Museum in Oslo.This set seems 

to be the one which Isak Saba, the first Sámi to be a Member of the Norwegian 

Parliament, offered to the Sámi Collections of the Ethnografisk Museum in 1906.The 

game is played with 15 pieces in each camp, one king and three 4-sided dice with 

three sides marked II, III, X and one blank (= 0).The board shows 15 short lines but 

no long central one.Two crosses on lines 4 and 12 suggest that there might have been 

originally two more ‘king’s sons’ as in the Petsamo and Utsjoki sets. April 2001 (Photo 

Angélina Vinciguerra, Paris). 

from what they could remember from their own youth. Roger Persson, a friend from that 

museum, told me that he and his colleagues sometimes used at lunch-break another 

reproduction of the same 1906 set for playing a game, confessing that they were often 

obliged to make up some of the rules. (6) 

Besides such ‘field interviews’, one may draw on museum collections, literature and 

vocabulary concerning the game. But, as most of the information collected does not 

concern the same place, the same groups or the same aspects of the game, one cannot 

even compare them, so that a lot of blank spaces still remain. 

The number of sáhkku sets now kept in museums is rather limited. (7) At the Norsk 

Folkemuseum in Oslo two series of pieces and two apparently complete sets are catalogued. 

However, one of these latter two is composed of pieces acquired in 1851 

(NFSA.0063) and a board (NFSA.0139) purchased in 1873! The other complete set has 

been exhibited since 1987 at the Varanger Samiske Museum in Varangerbotn as a longterm 

loan (fig. 2). This game seems to be the one which was presented to the Sámi 

Collections of the Oslo Ethnografisk Museum in 1906 by Isak Saba from Nesseby, the 

first Sámi to be a Member of the Norwegian Parliament. (8) 

The Tromsø Museum keeps one Russian Skolt Sámi set from Boris Gleb acquired in 

1951 (L.500) (fig. 3) and a series of 30 pieces for “sakkospill” of unknown origin (L.1171). 

At the National Museum of Finland, in Helsinki, we find some more material: 

• A wooden box containing the pieces of a pirccu game, said to have been collected 

in Inari or Kittilä (SU 1826:6). This set more accurately seems to be the one made by

36 


Fig. 3. Sáhkku game collected by Qvigstad at Boris-Gleb, in the coastal Skolt 

community of Paatsjoki (Petsamo territory, Russia).The board is much more compact 

(23 x 37 cm) and its surface is carved so to make squares of different depths.There 

are here two series of 8 men each, plus an extra short one, in the middle on the left, 

which could perhaps be one of the two “king’s sons” (Tromsø Museum). 

Fig. 4. Gameboard, pieces and dice made by an Inari Sámi in 1876.The board measures 

36.5 x 13.5 cm and shows 15 transverse lines.The pieces are: 15 ‘women’ (a) and 15 

‘men’ (c).There are crosses on the board marking the places of the king and the two 

‘king’s sons’.Three 4-sided dice were used, marked II, III, 4 and X (National Museum of 

Finland, SU 1826:6, in Itkonen 1941).


an Inari Sámi in 1876 and reproduced in Itkonen 1941 (fig. 4). The inside face of the 

lid may have been used as a gameboard, 

• A compact sized gameboard with 9 x 9 alternated black and white squares, coming 

from the Skolt forest community of Suenjel and curiously catalogued as pertsaloudi 

– literally “dice game” (SU 4922:189) (fig. 8). 

• One set wrongly catalogued as sahmatloudi (from shahmaty, Russian for “chess”) 

and said to have been collected among the Suenjel Skolt. It is a typical sáhkku game with 

a 75 cm long narrow board, three four-sided dice, 2 x 14 pieces, one king and two king’s 

sons (SU 4922:190) (fig. 7). 

• A series of 2 x 14 pieces of a pirccu game from Utsjoki, with one king, one king’s son 

(?) and three four-sided dice. (9) A hand-written illustrated note dated 1928 is joined to this 

series which was sold to the Museum in 1931 by Eliel Lagercrantz (SU 5126:14) (fig. 5). 

Fig. 5. Sáhkku pieces and dice collected at Utsjoki, Finland.The 30 pieces are from six 

different types: 1 of type a (king), 14 of type b (men), 3 of type c, 4 of type d, 7 of 

type e (=d, but black), 1 of type f.The 14 ones of types c, d and e might be ‘women’. 

A hand-written note joined to this set says however that the game is played with 15 

pieces in each camp.The three dice (g) have four sides: one blank (=0), the three 

others respectively marked II, III and X (National Museum of Finland, SU 5126:14, in 

Itkonen 1941). 

Even if Ernst Manker has quoted the game in several of his books (Manker 1954; 

Manker 1975), the Nordiska Museet in Stockholm, which he directed, does not keep any 

set in its rich Sámi collections. Neither the Karasjok Museum (Norway), nor the Ajtte 

Museum in Jokkmokk (Sweden) nor the Inari Siida Museum (Finland) have got any 

examples. 

Turning now to the literature, the earliest account we have dates from 1871 with the 

three pages the Norwegian Friis wrote on sáhkku in his book about Sámi mythology and 

folktales (Friis 1871: 164-167). (10) He gives the rules of the game as it was played in

38 


Finnmark, with illustrations showing the various game pieces and how they move (fig. 1). 

Three years later, the game was mentioned, with simple rules, in the catalogue of a Sámi 

exhibition in Göteborg (Sweden). 

Then we must wait for some 70 years to find something substantial in Lagercrantz’s 

and, moreover, Itkonen’s publications (Lagercrantz 1939; Itkonen 1941; Itkonen 1948). 

We discover there that the same game, with sometimes two more ‘king’s sons’, is not 

only known in Inari and Utsjoki, under the name of pirccu, but in the Skolt Sámi communities 

of the Russian border where it is called pertsa(loudi). 

In 1953, a four-page article from Yngvar Mejland tells us that sáhkku was actively 

played in Nord-Troms too, giving a short presentation of the game as it was played there 

(Mejland 1953). However, the major modern contribution to the knowledge of the game 

was given in 1999 by Peter Michælsen when he demonstrated the striking resemblence 

between the Sámi sáhkku and the Danish daldøs (Michaelsen 1999). 

The vocabulary used for naming the game and the various events it induces might 

certainly reveal some interesting information. Unfortunately, we have had no time until 

now for investigating it. What we know from the existing literature is that most of the 

words used when playing sáhkku are pure Sámi ones, which seems to indicate that the 

game belongs to an already old tradition. 

Geographical extension 

Before recording the various places where the sáhkku game has been played, we must 

say a few words about the Sámit. True aboriginal people of Lappland and fully differentiated 

from their Scandinavian and Finnish neighbours, they have until now maintained 

the strong original and rather monolithic culture which permitted their ancestors 

to survive in a severe arctic environment. This is all the more surprising because of their 

small number (about 70,000 now), dispersed over a vast territory (equivalent to about 

60% of France), itself cut by several borders and subject to the uses and laws of four 

countries (Norway, Sweden, Finland, and Russia). 

When looking back over their history, one is struck by their surprising flexibility 

which enabled them to face, however hard they might have been, the new conditions of 

life caused by the colonization of Lappland: the extinction of wild reindeer, taxes to be 

paid simultaneously to several states, borders moved or even closed, newcomers settling 

on their grazing lands, etc. Organized in small communities gathering a few families for 

mutual aid, they adapted themselves to what nature could provide them with: sea-fishing 

and light farming for the groups disseminated on the Norwegian coast (Sea Sámit), 

reindeer-breeding for most of the inland ones (Mountain Sámit), lake and river-fishing 

in the northern Finnish Utsjoki-Inari area (Fishing Sámit), and mixed small reindeerbreeding 

and river-fishing in the forest zones (Forest Sámit). The Russian Skolt Sámit, 

of Orthodox persuasion unlike the others who are Lutherans, form a separate group 

including both coastal and forest communities. 

Keeping that in mind, it can be said that sáhkku was mainly known in a narrow coastal 

fringe all along the Arctic shore from Nord-Troms (Tromsø area) in the west to the Russian 

Petsamo Peninsula in the east, in other words into Sea Sámi and Skolt communities.


With regard to Nord-Troms, Yngvar Mejland (Mejland 1953) writes that the game 

had been in use “until recent times” in various places. His great-grandfather, Nils Olsen, 

from the tiny Hauköya island at the mouth of the Kvænangenfjord, knew it since he was 

a child. He adds that sáhkku was known also in Nordreisa, between Lyngen and 

Kvænangen. This is confirmed by Edmund Johansen (Johansen 2000-2001), who says 

that it had been played in the Lyngenfjord area until the 1950s. 

As for Finnmark, the northernmost province of Norway, Edmund Johansen says that 

the game was the same in Sørøya Island west of Hammerfest. One of the sets kept by the 

Norsk Folkesmuseum in Oslo comes from Komagfjord, a small seaside place opposite the 

nearby island of Seiland. East of Cape North, one finds the sáhkku again in Laksefjord 

P o l a r C i r c l e 

Bergen 

Jaeren 

Thy 

DENMARK 

Fanø 

Trondheim 

NORWAY 

Mors 

Oslo 

SWEDEN 

Copenhagen 

Sørøya 

Hauköya 

Kunes Nesseby 

Nordreisa 

Neiden 

Utsjoki 

Boris-Gleb 

Lyngen 

Inari Suenjel Kola 

Nord-Troms 

Peninsula 

Kvikkjokk (?) 

Stockholm 

Bornholm 

Gulf 

of 

Bothnia 

Baltic 

Sea 

FINLAND 

Helsinki 

RUSSIA 

St. Petersburg 

0 100 200 300 400 500 

Fig. 6. Geographical expansion of the game.The thick gray line is the southern limit of 

the territory where the Sámit live.The gray squares are the places where sáhkku is 

known to have been played.The black triangles show the places where the Norwegian 

daldøsa and the Danish daldøs have been observed. (Map:Thierry Depaulis) 

km

40 


where Edmund lives. “In the beginning of the 50s and until 1966, he tells, the game was 

still played actively in Kunes, Bekkarfjord and Skjötningberg, three Sea Sámi villages in 

the fjord. Prior to 1975, the only road coming from Børselv was cut off for the whole 

winter, totally isolating the Nordkinn peninsula on the coast of which these villages are 

located. So, playing sáhkku helped people to fill in the long winter nights. The game 

was known on the coast among Sea Sámit but not among Mountain Sámit, who came 

up here with their reindeer for their summer camp.” (11) 

Continuing our way eastward, we reach the Varanger peninsula, on the west coast of 

which we know sáhkku to have been played in Trollfjord still in 1903. (12) On the south 

coast of that peninsula, Nesseby too has been an active center where the game was played. 

The set exhibited at the Varanger Samiske Museum in Varangerbotn (fig. 2) might come 

from there. 

Still following the Arctic coast on the southern bank of the Varangerfjord, we reach 

the Skolt Sámi area in Neiden, where some people remember the game still being played 

at the beginning of the 20th century. In a letter to the Varanger Samiske Museum dated 

April 25, 1996, Aud Mathisen said that her mother, Maila Mathisen, then 82 years old, 

recalled that, when she was still a child in Neiden, the game was called pirttsa and people 

said it was a Finnish game. 

Fig. 7. Gameboard, pieces and dice from the same coastal Skolt community of Paatsjoki 

(Petsamo territory, Russia) as in fig. 3.The board measures 75 x 14 cm and shows 15 

transverse lines.The pieces are of five types: 1 of type a (king), 1 of type b, 14 of type c 

(‘women’), 1 of type d and 14 of type e (‘men’).Types b and d seem to correspond to 

the ‘king’s sons’.The three 4-sided dice (f) are marked II, III, :: (=4) and X (National 

Museum of Finland, SU 4922:190, in Itkonen 1941).


From farther east over the Russian border, from the Petsamo territory which used to 

belong to Finland between the two world wars, we know three sets. One, in the Helsinki 

Museum (SU 4922:190), is a pure sáhkku game on a long plank (75 x 14 cm) (fig. 7). 

It is misleadingly catalogued as sahmatloudi (from shahmaty, Russian name for “chess”) 

and said to have been collected in Suenjel, a forest community broken up by the last war. 

The second example, in the Tromsø Museum (L 500), comes from Boris-Gleb, the main 

village of the coastal Pasvik community. Here the board is much more compact (23 x 37 

cm) and its surface is cut out to make up squares of different depths. But the king, the 

dice and the pieces (2 x 8) show it is a real sáhkku set (fig. 3). (13) The third specimen is 

just a gameboard kept in the collections of the Helsinki Museum (SU 4922:189) and catalogued 

as pertsaloudi. It too has a compact form, showing 81 squares (9 x 9), alternately 

black and white, and comes from Suenjel (fig. 8). 

Fig. 8. Gameboard collected in the Skolt inland community of Suenjel and catalogued 

as pertsaloudi (i.e. “dice game”) although it is more likely a shahmatloudi, i.e. a “chess 

game”. Note the basket pattern formed by the black squares (National Museum of 

Finland, SU 4922:189). (CAD drawing:Thierry Depaulis) 

Even if it seems that there has been an inversion about the names (SU 4922.190 is a 

pertsaloudi and SU 4922.189 is a shahmatloudi, the Skolt version of chess with black and 

white squares), (14) this third board is interesting because, with its compact shape and 

squares, it resembles the second board which has its sáhkku king, three dice and two series 

of eight sáhkku pieces. One may imagine that perhaps some of the Skolt could have used 

the same board for playing both games, a supposition which has to be confirmed. 

If we except the Suenjel games, all the places quoted until now are located directly 

on the coast, and, as we have said earlier, even the Karasjok Mountain Sámit, who

42 


camped near Kunes on their summer migration, did not play sáhkku. In Kautokeino, the 

other important community of Norwegian Mountain Sámit, the game is also completely 

unknown. It is obvious that it is mainly a seamen’s game, but it has been in use in some 

few places inland too. In Kvikkjokk, in Swedish Lappland, the word sahkku used to be 

known, referring to some kind of board for playing. From Finland, we have two of the 

sets kept by the National Museum in Helsinki. One was made by an Inari Sami in 1876, 

the other collected in Utsjoki in 1928 by Eliel Lagercrantz. In these two places, the game 

is called birccu or birccut – which means “die” or “dice”. It may be noticed that Inari 

and Utsjoki are both located in areas where, beside a few families of Mountain Sámit, 

most of the population is composed of Fishing Sámit. 

Let us come back to the sea and end up our survey at Spitzberg, where it seems that 

the sáhkku game might have been known too. In 1890 a Swedish archaeologist excavating 

in that island found a piece which he thought was a chess king. But any Russian 

piece who was there would have corrected this, saying that it was a sáhkku king. This 

story is by no means nonsense, for the Sea Sámit may perfectly well have frequented the 

archipelago. As the sáhkku king was generally carved in reindeer horn or cast from metal, 

it could have been the only piece that survived from a full set. 

Permanent features and local variants 

From the literature we understand there has been a certain number of local variants 

of the game, but we cannot describe each one in detail. (15) Our purpose here is just to 

try to distinguish between what is common and what is different in the various rules. 

• The dice 

Three dice are generally used, except in Nord-Troms where Mejland said they were 

only two. According to old people from Nesseby, who remembered the game, dice used 

to be cast into a bowl, not on a flat surface. Such dice were carved in wood like teetotums 

(26 x 26 x 42 mm in the Varanger / Nesseby set), with two of their end sides shaped 

like pyramids. All have four marked faces, and one of these always shows an “X” (sáhkku). 

The value of that “X” varies from place to place, either 1 or 5. In the Kunes (eastern 

Finnmark) rules, its value is 1, just like the “X” in the Norwegian daldøsa game from 

Jaeren and like the “A” in the Danish daldøs. In Friis’s description from Finnmark, it is 

worth 5 and it seems to have been the same in the game from Utsjoki. 

The values of the other marked faces vary too. In Friis’s version one is plain (=0) and 

the others are marked II (=2), III (=3), IIII (=4) and X (=5), the latter being coloured 

black. We find the same values in Kunes, Varanger and Utsjoki. (16) In Nord-Troms, 

however, they are 1-2-3-X; in Boris-Gleb and Inari, 2-3-4-X. 

In Kunes, as in Friis’s account, values of the dice could be used separately for moving 

several pieces and the king if it is owned. But they could also be accumulated for 

moving just one piece or the king. This was not allowed in Nord-Troms where each 

piece had to move according to the value of one die. 

In Nord-Troms and in Utsjoki, extra throws were granted to the player who threw 

one or several consecutive sahkku, an advantage which does not exist in the Kunes rules.


• The pieces 

The two series of gamespieces are generally called olbmát (men) and gálgut (women), 

even if Mejland calls them “soldiers” in Nord-Troms. Carved in wood, they are differentiated 

by their shapes. In the Varanger Samiske Museum set (fig. 2), the pyramids of 

the ‘men’ are topped with a small sphere while the ‘women’ have a crook-shaped form. 

Both might represent old-fashioned traditional Sámi hats: the cone-shaped bobbletopped 

one of the men and the women’s laggjo which had been forbidden by Laestadianists 

for it would have symbolized the “Devil’s horn”! Everywhere else they are much 

simpler, the ‘men’ usually are shaped as sharp-pointed four-sided pyramids, and the 

‘women’ like steep-roofed houses, with a notch on the top. 

The size of the pieces does not vary so much. In Kunes, both ‘men’ and ‘women’ 

measure ca 25 mm. In the Varanger set which we had in our hands, the ‘men’ are 38 mm 

high and the ‘women’ 33 mm. In comparison the king is 94 mm on a square base of 

26 mm width. 

In Inari as well as in Finnmark (Friis’s description, Kunes and Nesseby-Varanger), the 

pieces are 15 in each camp. However, Lagercrantz says that in 1928 in the Finnmark 

fisheries people used to play with 20 men and 20 women. In Nord-Troms, according to 

Mejland, there were only 12 of each. In the Skolt Paatsjoki set they were 14, and only 8 

in the Skolt Boris-Gleb example. (17) As far as the Utsjoki set is concerned, even if only 

2x14 gamespieces have been preserved, the hand-written note – be it by Lagercrantz or 

by Itkonen – joined to it does mention 15 in each side. 

• The board 

The standard board consists of a long narrow wooden plank on which 15 parallel 

lines are cut widthwise and sometimes one perpendicular middle line lengthwise. At the 

intersection of the long middle line with the eighth short line, i.e. at the center of the 

board, a cross is engraved: this is where the king stands at the beginning of the game. The 

size of the board varies from about 30 x 15 cm (Kunes and Inari) up to 56 x 15 cm 

(Varanger) and even 75 x 14 cm (Paatsjoki). 

Besides this type, which is the most common, the Skolt area shows two other models 

where the shape is much more compact (37 x 23 cm for the Boris-Gleb set at the 

Tromsø Museum). These two boards have no carved lines but they have squares (9 x9) 

that are differentiated either by their depth or by a black-and-white pattern. Even if the 

Boris-Gleb set is displayed with sáhkku pieces which seem to be of the same origin as the 

board, it is evident that these two boards were originally made for an other type of game 

more similar to draughts or chess. 

When comparing the various real sáhkku sets, one notices that in some of them 

(Nord-Troms and perhaps Utsjoki) the number of pieces per player is the same as the 

number of lines less one. Maybe it is this feature which led Itkonen to think the pieces 

were placed between the lines at the beginning of the game and not on the lines. Since 

the place of the king was marked at an intersection of the middle line and not between 

two lines he thought that, once arrived on this middle line, the men had to move from 

line to line. We saw that it was not the case in Kunes, where, wherever they are, the

44 


pieces always move from line to line. In Nord-Troms, however, where there are 12 men 

for 13 squares, we know from Mejland that, at the beginning, the place at the right of 

each player was empty: it was used for “activating” and playing the first piece. 

• The king 

The tallest and the most powerful of the pieces is the sáhkku king (gonagas), a pure 

Sámi addition to the daldøs(a)-type games. At the start of the play, it is placed in the center 

of the board which is marked with a crossed square called the “castle”. It has roughly 

the shape of a sharp 4-sided pyramid, either with simple notches on its ridges or 

showing several superposed tiers, some of them decorated with crosses – the sáhkku sign. 

An exception to this general type is the piece shown in Mejland’s article for Nord-Troms 

where the king is topped with a kind of bishop’s crook, a sign which could refer to the 

times when the Catholic church of Trondheim was still very influential there. 

Except in the Utsjoki variant where it may be killed, the king is a powerful piece, able 

to move in all four directions. So it is important to capture it. In Nord-Troms, Kunes 

and Utsjoki the king belongs to the first player who reaches the point where it is situated. 

According to Friis, in Finnmark, the player who first threw a sáhkku not only won 

the king, but was allowed an extra throw with the three dice. 

• The ‘king’s sons’ 

We know that king’s sons did exist from the two crosses which appear on several 

sáhkku boards (Varanger, Utsjoki, Inari and Paatsjoki), four lines on each side of the 

king’s place, but we have no idea about how they were used. From the differentiated 

shape of the only pieces kept in a museum (fig. 7), each one higher but apparently related 

in shape to the pieces of the two armies, we may guess that each player had his own. 

According to Itkonen, these would correspond to the ‘king’s sons’ of the Sámi tablut 

game from Frostviken (Swedish Lappland). 

The sáhkku king: god or devil? 

The social organization of the Sámit never had a king. There is one, however, in the 

Sámi game tablut, and moreover there are two ‘king’s sons’ in the Frostviken variants 

from Swedish Jämtland, just like in sáhkku. It has to be noted that this tablut king is 

called “king of the Swedes” and not “Sámi king”, which may find an explanation into 

the fact that tablut is not a genuine Sámi game but derives from the old Viking game 

hnefatafl. 

The sáhkku king and king’s sons are obviously connected with the tablut pieces, a 

proposition further reinforced by the fact that the pieces in the two games often have 

similar forms. As far as the king is concerned, it seems that the sáhkku king had a specific 

status. It could be in wood but more frequently in a harder material, such as reindeer 

horn, lead or walrus ivory, that kept it from decaying. It was decorated with several 

“Xs”, the sáhkku sign. In Kunes, it was topped with a four-pointed cap as the one that 

Finnmarks and Finnish Sámit wear, derived from the square chapka of the Russian 

Pomor merchants who traded along the coast.


Edmund Johansen (Johansen 2000-2001) remembers that, when he was still a child 

in Kunes, it was considered “bad” to play sáhkku. The children could play cards but 

sáhkku was strictly forbidden for it was said to be the “Devil’s game”. As a matter of fact, 

it seems that followers of Lars Levi Læstadius, the very rigorous Lutheran pastor who 

evangelized the Sámit in the 19th century, pronounced a kind of ban on this game. 

For Edmund Johansen the reason for such a ban was that the game of sáhkku alluded 

to old Sámi shamanic beliefs, against which Laestadianists fought actively. The crosses 

(sáhkku signs) with which the sáhkku king is decorated were symbols of the sun, which 

the old Sámit worshipped as the center of the world. “Our ancestors,” Johansen explains, 

“called it gonagas from the Norwegian word (konge, “king”) simply because, as they had 

no king, they had no word in their language for calling it. But this king may perfectly 

well have meant god with the signification the Sámit give to that word, i.e. a natural 

power, neither good nor bad, whith whom one has to deal anyway by making offerings. 

In our language sáhkku means ‘penalty’. You may know the Sámit often give things new 

names just to avoid the minister’s wrath, so they could say penalty while they really 

thought of offering.” 

In confirmation of such a theory there is at the Nordiska Museet in Stockholm the 

reconstruction of an old Sámi offering-place with a cross-marked wooden idol the general 

shape of which strangely resembles the sáhkku king (fig. 9). 

However pertinent our 

informant may be, it is obvious 

that the game of sáhkku has 

more than a smack of heresy. 

“One day,” tells Edmund Johansen, 

“a man found a leaden 

sáhkku king which he decided to 

use as a sinker for fishing. He 

said he knew it was the Devil, 

but he added that such a devil 

could make him lucky when 

fishing”. 

Fig. 9. Reconstruction of an 

old Sámi offering place with a 

cross-marked wooden idol the 

general shape of which looks 

like some sáhkku kings 

(Nordiska Museet, Stockholm).

46 


A still mysterious origin 

At the present stage of our research, it would be hazardous to formulate any theory 

about the origins of sáhkku. The only statement which makes sense is that the game is 

definitely not a true Sámi creation. The Sámi culture is so homogeneous that, if it had 

been so, the game would have been known in the whole Sámi area and not be limited 

to the fringes of the Arctic shores. As a matter of fact it seems that some Sámi groups 

invented a new game by introducing a king (and even two king’s sons) to a daldøs(a)-type 

game, both being evidently closely related. Such additions make the game much more 

tactical and attractive. But the concept of “king” has nothing to do with the Sámi tradition. 

As previously stated, when it appears in tablut, another Sámi board game from 

Sweden, sometimes also with two king’s sons, it is called “King of the Swedes” – not a 

Sámi king, and it fights the “Moscovites”. Tablut is thought to be one of the only survivors 

of the old Viking hnefatafl game. So the Sámi Arctic coastal groups to whom we 

seem to owe sáhkku would also have known tablut, although tablut was played in southern 

Swedish Lappland only; or they may have adopted the king and its sons directly 

from hnefatafl. 

So there are still a lot of questions to answer. As for how the original daldøs(a)-type 

game was introduced to Lappland, we offer here three possibilities, all each taking into 

account the remarkable present geographical concentration of the game in the Arctic 

shores of Troms and Finnmark: 

• The Viking track. This is of course the most satisfactory hypothesis since daldøs(a) 

has survived until now in western Denmark and southern Norway. The presence of the 

king and king’s sons in sáhkku, in relationship to the old Viking hnefatafl ones, seems to 

be another favourable argument. For centuries, but mainly around the 10th century, the 

western Viking ships made frequent sea raids up to the North where they might have 

taught local populations some kind of daldøs(a) game. If so, why is there – as far as we 

know – no survival of such a game on the 1,300 km-long coast between Jæren and 

Tromsø, even among the southern Sámit living there? And why would only the Sea Sámit 

from Troms and Finnmark have maintained its tradition? Is that just because they invented 

a mixture of hnefatafl and daldøs(a) which made the game much more exciting? 

• The Pomor track. From ca 1740 until the beginning of the 20th century, Russian 

merchants from Archangel called Pomors used to come every year to the coast of 

Northern Scandinavia on board heavily-laden sailing ships carrying flour, timber and 

hardware. They exchanged their freight for dried fish and other local productions with 

the Sea Sámit who had no other source for supplies. Trading was done in “Pomor”, a kind 

of Nordic Creole made of Russian, Norwegian and Sámi words, which was perfectly 

understood from Tromsø to Murmansk. Such a regular trade induced the Sea Sámit to 

adopt several cultural features from their yearly visitors, among other things their curious 

four-pointed caps. The daldøs(a)-type game could have been one of those, but, as far 

as we know, there is no evidence that such a game has ever been played in Northern 

Russia.


• The Kvaen track. If sáhkku did not come by sea then it could have been brought 

by the Kvaen, Finnish emigrants who, around the year 1800, came in large numbers to 

settle in the ends of the fjords of Nord-Troms and Finnmark. In 1913, they numbered 

no less than 7,000 people in these two provinces. In several areas of eastern Finnmark, 

they even formed the majority of the population. Among the main places where they 

concentrated are Skibotn, east of Tromsø, Nordreisa, Inari, Utsjoki, Vadsö, Bugøyfjord 

and Neiden, the latter three on the Varangerfjord, all precisely spots where we know 

sáhkku was played. Besides such geographical agreement, we know from a Neiden 

woman’s testimony that, at the beginning of the 20th century, the game was said to be 

a Finnish one. The fact that, in the Finnish language, the word sakko has the same “penalty” 

meaning as its Sámi equivalent would also support the idea of a Finnish origin. But, 

just as with the Russian track, we have so far never heard that this kind of game was 

known anywhere in central or southern Finland. 

This short analysis shows how difficult it is to explain the presence of such a ‘tâb’ 

game such a long way from its original area. Even if sáhkku appears to be derived from 

a daldøs(a)-type model, it remains to be demonstrated how this model reached the 

Fenno-Scandic area. 

References 

Drake, Sigrid 1918. Västerbottenslapparna under förra Hälften av 1800-talet. Uppsala. 

Friis, J.A. 1871. Lappisk mythologi, eventyr og folkesagn. Christiania (Oslo). 

Itkonen, Toivo Immanuel 1941. Die Spiele, Unterhaltungen und Kraftproben der 

Lappen, In: Journal de la Société finno-ougrienne, 51,4: 1-133 (§106. Würfel. § 107- 

108. Die Brettspiele). 

Itkonen, Toivo Immanuel 1948. Suomen Lappalaiset vuoteen 1945, I-II. Poorvo-Helsinki. 

Johansen, Edmund 2000-2001. Personal interviews of Edmund Johansen in Kunes, eastern 

Norwegian Finnmark, May 2000 and April 2001. 

Keyland, Nils 1921. Dablot prejesne och dablot duoljesne. Tvänne lappska spel från 

Frostviken, förklarade och avbildade, In: Etnologiska Studier tillägnade Nils Edvard 

Hammarstedt, 3/3, Göteborg: 35-47. 

Lagercrantz, Eliel 1939. Lappischer Wortschatz, Helsinki: No. 4945. 

Manker, Ernst 1954. Les Lapons des montagnes suédoises, Paris. 

Manker, Ernst 1975. De åtta årstidernas folk. Göteborg. 

Mejland, Yngvar 1953. Sakko, In: By og Bygd (Oslo), 8: 00-00. 

Michaelsen, Peter 1999. Daldøs og Sakku, to gamle nordiske spil med fjerne slaektinge, 

In: Ord & Sag, 19: 15-28 

Schefferus, Johannes 1673. Lapponia, Frankfurt. 

Notes 

1. The word Lapp is a nickname given by the first Scandinavian settlers who met the Sámit. In 

the old northern dialects, it meant something like “rags”. Sometimes still called Finner in 

Norway – Finner for “Finnish”, although these two populations differ quite a lot – the Sámit

48 


(Sámi in the singular) have now recovered, at least in the Fenno-Scandic area, the name they 

have always given to themselves. 

2. See Thierry Depaulis, “Jeux de parcours du monde arabo-musulman”, this volume. 

3. In Sámegiella, the Fenno-Ugrian language of the Sámit, the “á” is pronounced like a in the 

English “sad”, and the “u” is pronounced . So, the word sáhkku (meaning “penalty”) 

must be pronounced . The letter “c” being the equivalent of “ts”, bircu (die) and 

birccut (dice) – two other names given to the game in the Utsjoki-Inari Finnish area, have to 

be articulated and , which sound rather like the word percc’ (die), 

pronounced , used by the Skolt Sámit for naming the game. 

4. Michaelsen 1999 and this volume. 

5. Of course, these numbers do not appear on the board. They are simply given here to explain 

how the pieces move. 

6. In April 2001 Roger Persson told us that the rules they used were in fact those published by 

Friis 130 years ago. When playing a real game, these rules appeared rather incomplete. 

7. We want to specially thank here Mrs. Nina Puurunen, Curator of the National Museum of 

Finland in Helsinki; Mrs. Elisabeth Brundin, Curator of the Nordiska Museet in Stockholm; 

Mr. Leif Pareli, Curator of the Sámi Collection at the Norsk Folkemuseum in Oslo; Mr. Terje 

Brantenberg, from the Tromsø Museum; Mr. Roger Persson, from the Varanger Sámiske 

Museum in Varangerbotn – all of whom provided co-operative and efficient help. 

8. According to curator Leif Pareli, from Oslo, there is an irregularity in the numbers in the catalogue 

so that set could belong to another collection acquired in 1906. However, he says, 

“I find it more likely that this game was actually acquired through Isak Saba”. 

9. Peter Michaelsen thinks that the so-called king’s son in fact is the missing king of the Utsjoki 

game (SU 128:14), for it is identical with the drawing in Itkonen 1941: fig. 34. It would have 

been misplaced by the museum. 

10. Two centuries earlier, Schefferus wrote that similar dice were used by the so-called Lapps, but 

he did not mention any board game related to them. 

11. Such a strictly coastal distribution looks very similar to the one we observed for the French 

aluette, an old seamen’s card game played from Cherbourg to Bordeaux on a narrow strip all 

along the coast and the banks of the Loire river up to Orléans (see Alan Borvo, Anatomie d’un 

jeu de cartes: l’aluette ou le jeu de la vache, Nantes, 1977). 

12. Konrad Nielsen, Lapp Dictionnary (s.v. “Sáhkku”). 

13. However, one cannot dismiss the idea that the Tromsø Museum photograph shows an arrangement 

made out of an incomplete sáhkku set (9 ‘women’, 8 ‘men’, 1 king, 3 dice) “artistically” 

set up on a draughtsboard! (Editor) 

14. Such an inversion is confirmed by the fact that the real Skolt sáhkku game wrongly named sahmatloudi 

in the catalog of the Museum (SU 4922:190) is said to come from Suenjel, although 

Itkonen, who himself collected these games in 1913, presents this one as coming from Patsjoki 

(Itkonen 1941). 

15. The only rules we are sure of are the ones of the Kunes variant which we give hereafter. 

16. In Utsjoki too, the X-marked side was coloured black. 

17. The latest set seems to be a shorter version of sáhkku played on a board made for an other 

game. (But see Editor’s remark note 13.)


Appendix 

Rules of sáhkku 

(Kunes variant, eastern Norwegian Finnmark) 

The rules we give here have been collected directly from Edmund Johansen, a Sea 

Sámi from Kunes who had been actively playing the game there until the 1950s. They 

have been written down after we played several games together in April 2001. As we 

wanted to provide scholars and players with a complete useful account, some parts of this 

text necessarily repeat what has been said earlier in the main article. We apologize for it. 

Material 

• One gameboard made of a small wooden plank (30 x 15 x 1 cm) on which are 

carved 15 parallel lines widthwise and one perpendicular central line lenghwise. Just at 

the center, at the intersection of the long central line with the eighth short line a cross 

is engraved: it is where the king stands at the beginning of the game. 

• Fifteen ‘men’ (olbmát in Sámi), ca 25 mm high square wooden pieces, cut in the 

shape of long 4-sided pyramids. 

• Fifteen ‘women’ (gálgut in Sámi), square wooden pieces of the same size as the previous 

ones, but cut in the shape of steep-roofed houses with a notch on the top. 

Fig. 10. Sáhkku lesson given to the author by Edmund Johansen in Kunes (eastern 

Norwegian Finnmark) in April 2001. Edmund is considered the best expert on the 

rules of the game (Photo Angélina Vinciguerra, Paris).

50 


• One king (gonagas in Sámi), a ca 45 mm high square piece made of wood, reindeer 

horn or cast tin. It shows four superposed tiers, the second ornamented with four crosses 

(the sáhkku sign), and the fourth cut as a four-pointed cap. 

• Three dice (birccut in Sámi), cut from wood and longer than the usual ones. They 

have only four scoring sides, the two ends forming 4-sided pyramids. The values marked 

on the sides are: X (sáhkku = 1), II (=2) and III (=3). The last side is blank (=0). Usually 

the dice are thrown into a wooden bowl. 

Object 

The game is for two players. The aim is to “kill” all the opponent’s pieces or to prevent 

him from “activating” whatever pieces he has left. 

Beginning 

The board is placed between the two players with its long sides in front of them. 

Each chooses the side with which he will play – the ‘men’ or the ‘women’. He puts them 

in line on the side of the board in front of him, each man on the end of one of the fifteen 

parallel lines. The king is placed on the central cross. 

In order to determine who begins, each player throws a single die. The first to get a 

sáhkku (the X-marked side) plays first. 

At the beginning, the pieces are considered as locked in their places. To “activate” any 

one of them, a player must get three sáhkku in three successive throws using the three 

dice. After each throw the one or two sáhkku points obtained are put aside and the 

remaining die or dice are thrown again. Each player tries in turn. As soon as he has got 

three sáhkku, the player may “activate” some of his pieces: either (1) the first three on his 

right, moving each one jump, or (2) only the first, moving it three jumps, or (3) the two 

first moving one one jump and the other two jumps. 

Moving the pieces 

Throughout the game, no piece may leave its original place without being activated 

by a sáhkku, which will allow it to move one point (or line). The pieces must follow a 

track which leads them to meet the opponents, without any possibility of moving backward. 

The figure below will help the reader understand how they have to move. The 

pieces initially located from 01 to 15 move from the left to the right of their owner, then 

from 15 to 16, from 16 to 30, from 30 to 31, from 31 to 45, from 45 to 16, from 16 to 

30, and from 30 to 01. The opponents move similarly, but in the opposite direction: 

from 45 to 31, from 31 to 30, from 30 to 16, from 16 to 15, from 15 to 01, etc. 

Each player throws the three dice in turn. He may use the values given by them either 

for activating or moving one, two or three of his pieces. But the move of each piece must 

correspond to the exact numbers given by one, two or three dice. For example, a throw 

of X-III-blank (total = 4) does not allow to move two men two lines each: the player must 

move one man one line (for the X) and another one three lines (for the III). 

It is possible for a player to put one or several pieces in a place already occupied by 

one or several of his own pieces, with the risk of having them all killed if the opponent 

comes to this very place. Pieces may also jump over one or several pieces, of either side.


31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 

30 29 28 27 26 25 24 23 22 21 20 19 18 17 16 

01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 

Killing the opponent’s pieces 

When a piece comes to a place already occupied by one or several of the opponent’s 

pieces, the latter is or are killed and permanently removed from the board. Only already 

activated pieces may be killed. Killed pieces cannot be re-entered. 

Fig. 11.The set used by Edmund Johansen is a modern copy of those with which 

Kunes Sea Sámit used to play until the 1950s.The board is 30 x 14.3 cm, engraved 

with one central and 15 transverse lines.The pieces are 15 ‘men’ (background), 15 

‘women’ (foreground) and one king.Three 4-sided dice are used. April 2001 (Photo 

Angélina Vinciguerra, Paris).

52 


The king’s part 

Although the ordinary pieces are obliged to follow their own track forward, the king 

can move in all four directions. Set at the beginning of the game in the middle of the long 

central line (point 23), it will belong to the first player who will reach that place. 

However, throughout the game, it may be captured by an opponent’s piece which arrives 

where it stands. In that case, the king is not removed from the board but then plays with 

the player who captured it. 

The player who owns the king may use it, on his turn, with the values given by the 

dice. He may choose not to move it and move some of his pieces, or to move it alone 

using the total of the three dice, or to move it simultaneously with one or two of his other 

pieces. The ownership of the king does not allow any extra throw of the dice. 

The king kills the opponent’s pieces just like the ordinary pieces do. As it may move 

forward, backward and sideways, it is a very powerful and dangerous piece. Its moves, 

however, must be in straight line. If it stands on the long central line, for example, the 

player who owns it will need a sáhkku (X = 1) for killing an opponent’s piece located just 

in front on the side of the board. With a II (= 2), the king cannot make a right angle to 

kill a piece located on the side of the board but on the next line. 

Jamming on non-activated pieces 

When an opponent’s piece or, more frequently, the opponent-owned king comes to 

a place situated just on the right of a player’s non-activated pieces, these are jammed in. 

They will remain so as long as this piece or the king stay there. In that case, if all the activated 

pieces are killed, that player loses the game.

Sáhkku, The “Devil's Game” / Alan Borvo - Board Game Studies

Create successful ePaper yourself

Delete template?

Save as template?